1. Field of the Invention
The present invention relates to a region detection device and a region detection method which detect, for example, the entire region of an attention person from a still image extracted from a moving image, an image processing apparatus and an image processing method which generate a composite image including, for example, the detected entire region of the attention person, and a non-transitory computer-readable recording medium storing a program.
2. Description of the Related Art
In recent years, many families have captured moving images. A captured moving image is likely to include the best shot scene (a scene which appropriately shows the motion of a person captured in the moving image) which (which is difficult to capture) is not captured in a still image, such as a scene captured at the moment when a child blows out candles on his or her birthday. In some cases, the moving image includes a scene in which the motion of the person is small, a scene which is of low importance, a scene which has bad composition, and a scene with low image quality.
Therefore, it takes a lot of time and effort to detect the best shot scene from the moving image and to extract the best shot scene as a still image.
There are JP2009-75802A and JP2010-109592A as the prior art documents related to the invention.
JP2009-75802A relates to a person action search device which can rapidly reproduce moving image data from a recording position where a person is recorded. JP2009-75802A discloses the following technique: when a person in a captured image is recognized, a representative image of the recognized person is extracted; and a tracking line, which is a movement locus for the period from the appearance of the virtual center of gravity of the image of the person in the captured image to the disappearance of the virtual center of gravity from the captured image, is combined with the representative image to generate a bookmark image.
JP2010-109592A relates to a technique which extracts a representative frame directly indicating an image included in moving image data from the moving image data. JP2010-109592A discloses a structure which extracts one or a plurality of representative frames that indicate content (image) in a predetermined time section of the moving image data from the section. In addition, JP2010-109592A discloses a structure which extracts, as a representative frame image, a frame image in which an evaluation value output from a face state determination unit is the maximum.